0

I am running OLS regressions on time series data (interest rate swap spreads). I have run the regressions with Newey-West standard errors to deal with both autocorrelation and Heteroskedasticity.

My question is: is this enough to ensure my model is appropriate from a hypothesis testing standpoint? Or do I need to use first differences as well (since the data is not stationary)?

Thank you.

Eric
  • 1
  • Thanks for your help. – Eric Jun 08 '18 at 13:11
  • First of all, this is a question suited more for cross-validated, the statistics-oriented stack exchange, and second of all, this is an incredibly broad question to ask. You would be better served by going back to the theory and trying to understand more for yourself. – Coolio2654 Dec 04 '18 at 20:19

0 Answers0